© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 9336
    Prev
    Next
    INDEX
    Explanations

    numeric values

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    168
    +0.12
    0.6%
    140
    +0.10
    0.5%
    255
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.12
    0.03
    168
    +0.10
    0.03
    26
    +0.10
    0.02
    Negative Logits
     Franch
    -0.71
     Sussex
    -0.65
     Sparkle
    -0.64
    Ĥª
    -0.63
     CSI
    -0.60
    ãĥ¯ãĥ³
    -0.60
     Magnus
    -0.59
     Leilan
    -0.58
     Despair
    -0.58
    ãĥ¢
    -0.57
    POSITIVE LOGITS
    arette
    1.21
    gers
    1.18
    gered
    1.15
    arettes
    1.12
    rams
    1.09
    raine
    1.08
    abyte
    1.08
    raphic
    1.08
    roups
    1.07
    lia
    1.07
    Activations Density 0.071%

    No Known Activations